(old)Puppy Linux Discussion Forum

Hello Everyone

The Kennel is such a nice place to play, especially with all the new pups coming out like Precise, Slacko, Dpup and many more yet to come.

I like precise even though with the pae kernel I will not be able to use it very will on my machine. but that is ok will test even though things don't work so well.

I have tested puppies that I only use once and than that is it. Like to test new ones who others have done to help keep things going. I don't do a lot of different things with all the programs only use just a few so that is what I test for if what I use works. than let them now if it does or does not.

Your right on puppyluvr

Cheers

RC1 is running very well on my MSI netbook.

It's nice to finally have a woof build with an mplayer
that can play full screen.

It would run even better on the 1.6 Gb Pentium M
processor in my Inspiron 700M.
Unfortunately it won't run with this horrible BUG known as
a PAE kernel.
It's downright ridiculus that a machine that will easily run
Win7 or Mac SnowLeopard is off limits to a Puppy version
just so a few wealthy weirdos can use insane amounts of
ram.

The whole PAE notion is a total antithesis of the original
stated purpose of Puppy's creation.

+1

My hardware goes up to 16 Gb of ram and I don't run a PAE kernel..

James C wrote:+1

My hardware goes up to 16 Gb of ram and I don't run a PAE kernel..

I probably should have added a smiley to the "wealthy
weirdo" comment; I depend on the rich for survival .

Aside from machines that can't boot PAE; no machine that
uses Intel 945 graphics can use the extra ram .

Win 8 is planning on using PAE, and many people are "up
in arms" about it.

+2

PAE support is a good option to offer for computers that support it, but what are the computers Precise Puppy is designed for? Just those computers?

The Linux kernel includes full PAE mode support starting with version 2.3.23,[12] enabling access of up to 64 GB of memory on 32-bit machines. A PAE-enabled Linux kernel requires that the CPU also support PAE.

I have never seen Puppy use 1GB of memory in normal operation.
Seems to be a feature that lets you see the total ram on a computer, but Puppy will never need to use it.

The ffmpeg on the precise puppy has no capability to encode MPEG4-AVC(h264/aac).
I prepared libav-extra-0.8.3-up.pet to enhance extra encoding audio/video.
http://shino.pos.to/party/bridge.cgi?puppy/precise/

The pet contains the libraries from the next 11 ubuntu packages.

libavcodec-extra-53_0.8.3ubuntu0.12.04.1_i386.deb
libavdevice-extra-53_0.8.3ubuntu0.12.04.1_i386.deb
libavfilter-extra-2_0.8.3ubuntu0.12.04.1_i386.deb
libavformat-extra-53_0.8.3ubuntu0.12.04.1_i386.deb
libavutil-extra-51_0.8.3ubuntu0.12.04.1_i386.deb
libdirac-encoder0_1.0.2-4build1_i386.deb
libopenjpeg2_1.3+dfsg-4_i386.deb
libpostproc-extra-52_0.8.3ubuntu0.12.04.1_i386.deb
libswscale-extra-2_0.8.3ubuntu0.12.04.1_i386.deb
libvo-aacenc0_0.1.1-2_i386.deb
libvo-amrwbenc0_0.1.1-2_i386.deb

NOTE: Use 'libvo-aacenc' instead of 'libfaac' for the ffconvert MPEG4-AVC encoding.

I would like to test Precise RC1 but unfortunately, having just downloaded it, I have discovered that the PAE kernel does not run on my machine (a Fujitsu Siemens Amilo laptop with Pentium M processor)

I'll do my best to try it from a usb stick on my partners desktop when I can push my way onto that sometime.

Guys, this is based on Ubuntu, which now has a policy of only issuing PAE versions, due to most PCs being able to run it. I think Barry is simply following suit. There are lots of great non-PAE puppies still being developed. I use Luci-001.005 (non-PAE) from Playdayz on a laptop with 3gb ram - it's an update of 529 & it's great. Check it out.

shinobar wrote: I prepared libav-extra-0.8.3-up.pet to enhance extra encoding audio/video.

NOTE: Use 'libvo-aacenc' instead of 'libfaac' for the ffconvert MPEG4-AVC encoding.

Tried it on the immediately previous non-PAE Precise and libvo-aacenc worked OK on a quick test. However this libav/ffmpeg version seems to have brought back an old well-documented problem when using libxvid where you get the error message "Invalid pixel aspect ratio 0/1". EDIT: Though I have still to try it on the latest Precise RC1, which I shall do later.

The common workaround is to not use libxvid but instead: "-vcodec mpeg4 -vtag xvid", however that isn't apparently quite as good as libxvid without extra parameters also being added. The alternative seems to be to use libxvid but with the -aspect parameter in a -vf construct. For example: but that is a real pain because it needs to be recalculated for each resolution used. As things stand, however, I'll just use the "-vcodec mpeg4 -vtag xvid" simple alternative (which does work) even if it is slightly inferior to libxvid .

I haven't had a chance to test the new Precise yet owing to PAE kernel not working on the computer I usually use.

Just wanting to note that the previous version doesn't recognise video container ogv, which is the recommended extension for Theora video with vorbis ogg audio. So I wish to request that be fixed if not already done so since it is one of the optional containers used in pAVrecord Video/webcam recorder for which I'm producing a version to work on Precise.

Barry,
Please be sure to include SeaMonkey 2.13.1.

SeaMonkey 2.12.x has a nasty SVG bug:
Bug 790144 - SVG "stroke" hangs SeaMonkey 2.12.x and Firefox 15.0.x. Regression.
https://bugzilla.mozilla.org/show_bug.cgi?id=790144

That hang seems to happen with any SVG file
containing any kind of line that also has the
"stroke" thingy specified, which is needed
for the line to appear!

Therefor, I would classify SeaMonkey 2.13.1
as an "Essential bug fix".

Thank you very much,
Eddie Maddox

Report Video 1.3 - Wed 17 Oct 2012 on Precise Puppy 5.3.96 - Linux 3.2.29 i686

Chip description:
2.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
oem: Intel(r)852MG/852MGE/855MG/855MGE Graphics Chip Accelerated VGA BIOS
product: Intel(r)852MG/852MGE/855MG/855MGE Graphics Controller Hardware Version 0.0

X Server: Xorg Driver used: intel
X.Org version: 1.11.3
dimensions: 1024x768 pixels (270x203 millimeters)
depth of root window: 24 planes

Generally all seems good - very nicely setup - (fonts etc)
FYI - sys-report.

The actual choice of the coulour-sheme might attract some, but in my opinion it's a little usability-nightmare.

As an example please try to find the marked text in the attached pic. For my eyesight it's nearly impossible.

aragon

Hi Guys:

My initial report, http://murga-linux.com/puppy/viewtopic. ... 725#658725 indicated the absence of problems even though my computer has Ati Radeon 3100 Graphics. In the main, this is still correct. However, I noticed that sometimes the colors used in dialogues made them difficult to read. See attached. So I decided to experiment to find out if any of the amd_fglrx drivers might work, even though they weren't built for this kernel. [In Exprimo, the driver built under the 3.4.2 kernel actually worked with the 3.6.2 kernel]. Neither amd_fglrx-12.8-precise-k3.2.28 nor k3.2.29 worked. Both rendered the OS sort of unstable: to activate the driver, you have to exit X and type aticonfig --initial; exiting X hung the system. Closest was the aforementioned fglrx driver from Exprimo. It just sometimes reduced graphics to a crawl.

@ ozsouth: There are obvious advantages to having a Puppy which is able to utilize Ubuntu application binaries. But Puppy isn't Ubuntu. We are not trying to compete with Windows7 by providing an alternative fancy and equally bloated system. Puppy's advantage, whether employed on modern, state-of-the-art, or older, resource-limited computers is its low-ram & CPU usage and the ease with which applications can be initiated. With the possible exception of hardware drivers, there are no applications which if built under PAE-enabled kernels won't run on non-PAE enabled kernels. The only advantage a PAE-enabled kernel has over a non-PAE enabled kernel-with-smp is that the former can make use of more than 4GB of RAM. That advantage is of little significance in an OS designed to function with as little as 128 to 256 Megs of RAM, and fly with 1Gb of RAM. As far as I know, there are no computers which can manage more than 4Gbs of RAM which aren't also able to handle 64-bit operating systems. Comparison tests between software run under PAE-enabled and non-PAE-but-smp-enabled computers reveal at best an imperceptible (mili-second) advantage to the former in some cases, and even a disadvantage in others, while 64-bit systems have a clear advantage. So while it makes sense to build 64-bit systems, and 32-bit-smp-enabled systems, it is unlikely that there are a significant number of computers users which actually would benefit from a PAE-enabled system. And certainly not using a Puppy.
T2 builds such as Quirky/Racy or even Carolina are not substitutes for a Ubuntu-binary-compatible Puppy. As building a non-PAE Ubuntu compatible Puppy is possible --it's already been done a couple of times with the 3 Series Kernels-- and as Barry K is an innovator rather than a follower, as I said before I hope he reconsiders concentrating his efforts on a PAE-enabled kernel.

mikesLr

hi mikeslr,

i believe your radeon 3100 card requires a different driver , amd now have two drivers, and the one i compilied and is in the repo, is the wrong driver for you. i didn't realise there where 2 drivers for amd or i would have mentioned it on my post. http://support.amd.com/us/gpudownload/l ... linux.aspx

the driver is easy enough to compile it yourself, a .pet would be nice though

Yep. There is at least 12.6 and 12.8 for different graphics, depending is it for desktop or laptop/dual graphics.
Nowadays it would be useful to check the needed version from this page:
http://support.amd.com/us/gpudownload/Pages/index.aspx

First you need to know what ati graphics you have.
# lspci |grep VGA
00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 12)
01:00.0 VGA compatible controller: ATI Technologies Inc Redwood [Radeon HD 5600 Series] (rev ff)

Yeah....I have dual graphics, makes things really interesting in Puppy world.

I believe mikeslr would have ATI HD 3100, and possibly he uses desktop, lets put those specs in the above link and Display results: AMD Catalyst™ 12.6 Proprietary Linux x86 Display Driver

...and my ATI needs: AMD Catalyst™ 12.8 Proprietary Linux x86 Display Driver

Yeah....different package. It really is useful to check the needed package first before downloading and testing it.

By the way....If you rename the package to ati.run, you need just ./ati.run in X, choose Generate distribution spesific package, choose slackware, run it, grab the created /fglrx-install.f1XXXX/packages/Slackware/_working_directory_ ie the working directory and copy it safe with all stuff inside, rename the working directory as suitable, decompress the created fglrx.ko.gz in its folder and dir2pet the renamed working directory. And do that before you exit from the gui, otherwise all the extracted stuff will be deleted.
It might even work.

Test machines are both older Dell towers, one a Precision 330 and the other a Dimension 2350. Both have P4 cpus at 1.7 and 1.8 ghz. Both have 768 meg RAM. One has a Intel 82845G graphics controller and the other a Nvidia PCI card. Precise seems to correctly identify the video hardware
but .... the old black screen at startup occurs randomly on both machines, and it happens fairly often. The GUI is still useable when this occurs since the toolbar at the bottom is ok (menu is available, etc.) Then sometimes restarting X works. Sometimes trying Vesa doesn't work at all (Completely
white screen).

RC1 is still sorta flakey establishing wireless connection as well. I've managed to get it working by trying SNS repeatedly and when actually
connected to create the .2fs file. After that and reboot, it may or may not restart connected but one more connect finally does the trick. (It would be nice if Puppies had a green "connected ok" icon).

No such problems with other Puppies. Latest Lucid and Luci in particular work very well on these machines. Best of all is Luci which seems to be a bit faster that Lucid.

Art

ozsouth wrote:Guys, this is based on Ubuntu, which now has a policy of only issuing PAE versions, due to most PCs being able to run it. I think Barry is simply following suit.

Not all buntu's....... Lubuntu 12.04 32-bit or Xubuntu 12.04 32-bit; both of these use a non-PAE kernel by default.

http://askubuntu.com/questions/117744/h ... ures-not-p

Fresh manual frugal install on the Athlon XP box. No real problems.......display,sound and internet working on initial boot.

# report-video
Report Video 1.3 - Wed 17 Oct 2012 on Precise Puppy 5.3.96 - Linux 3.2.29 i686

Chip description:
0.0 VGA compatible controller: NVIDIA Corporation NV18 [GeForce4 MX 440 AGP 8x] (rev c1)
oem: NVidia
product: NV18 () Board Chip Rev A2

X Server: Xorg Driver used: nouveau

X.Org version: 1.11.3
dimensions: 1440x900 pixels (380x238 millimeters)
depth of root window: 24 planes


...the above also recorded in /tmp/root/ as report-video,
and archived with xorg.conf and Xorg.0.log as report-video-full.gz
# glxgears
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
1393 frames in 5.0 seconds = 278.499 FPS
2164 frames in 5.0 seconds = 432.653 FPS
2103 frames in 5.0 seconds = 420.447 FPS
2167 frames in 5.0 seconds = 433.352 FPS
1993 frames in 5.0 seconds = 398.483 FPS
2036 frames in 5.0 seconds = 407.129 FPS


-Computer-
Processor : AMD Athlon(tm) XP 2400+
Memory : 1034MB (205MB used)
Operating System : Unknown distribution
User Name : root (root)
Date/Time : Wed 17 Oct 2012 03:23:59 PM CDT
-Display-
Resolution : 1440x900 pixels
OpenGL Renderer : Mesa DRI nv18 x86/MMX+/3DNow!+/SSE
X11 Vendor : The X.Org Foundation
-Multimedia-
Audio Adapter : VIA8233 - VIA 8235

precise 5.3.96 (rc1) - first look..

Needed "puppy radeon.modeset=0" bootcode for useable display on my hardware - if bootcode not entered, display goes illegible (repeated gray/green scrunched lines) after "Loading kernel modules...".

primary p/s2 mouse correctly recognized and useable in initial quickset (dropdowns ok as well)

Default primarily-white backdrop looks clean, but initial help page has no visible border on it. Neon icon theme is a nice change.

Same minor errs as in 5395. FFConvert and ffconvert in rox OpenWith. Pmount tabview incorrect with usb in optical tab and optical in second floppy tab.

Tried my personal sfs with the on-the-fly loader .. nice to have some other wms and etc just to see my preferred 'usual' for a comparison. Afterstep's gtk-handling is almost always more legible and less prone to the too-subtle vagaries in some apps (it's somewhat rare for everything to be be nicely contrasted all the time, some fields suffer and so on ). Personally I'm of the opinion that fltk and qt may be less problematic than gtk at times.

____________
test on: i686 machine (desktop)
processor - Pentium 3 @1GHz
RAM - 512M
vidcard0 - RV280 (aka ATI Radeon 9200), AGP
vidcard1 - not used, but also RV280 (switched off in BIOS - aka Radeon 9200SE), PCI (old-style)
soundcard - emu10k1 (aka Creative Audigy 2 ZS or SB0350)
boot/save method - liveCD multisession (cd-rw media)
limitations - no printer, no internet connection